🎙 "We're Lifting Barriers For Outside Users; If You Allocate Capital in a Vaccum, It's Kind of Meaningless:" Terra's Do Kwon

In this week’s episode, we interview Do Kwon, cofounder of the Terra ecosystem, which includes a payments network, a stablecoin, a lending protocol and a synthetic assets platform. We talk about how he was able to achieve what most crypto companies can’t: Get a broad non-crypto native user base. Terra’s Chai payments system, which uses the Terra stablecoin, has about 2.6M active users in Korea. Kwon says there’s so much money to be made without bringing new people in, that many developers just settle with looking inwards in crypto.

Kwon explains how the Terra stablecoin is able to hold its peg, and how its use drives value back to the Luna token. We also talk about Anchor, Terra’s lending platform, and how it’s able to offer 20% fixed-rate yields.

Kwon dives into the vision behind Mirror, a synthetic assets trading platform. The goal is to give people anywhere in the world access to invest in US equities. He says US companies lead the world in terms of innovation but it’s very hard for people outside of the US to benefit from their growth.

At a time when much of crypto is focused on its equivalent of “Wall Street” or wealthy investors and speculators, Kwon wants to focus on making products for “Main Street.” Otherwise, he says, capital is just allocated in a vacuum instead of where there is actual production taking place.

The podcast was led by Camila Russo, and edited by Alp Gasimov. Transcript was edited by Owen Fernau and Dan Kahan.

🎙Listen to the interview in this week’s podcast episode here:

You’re a paid subscriber, which means you get the full transcript below. Subscribers also get exclusive access to The Defiant’s Discord chat for the community, here’s a new link to join.

Share The Defiant

🙌 Together with:

• Zerion, a simple interface to access and use decentralized finance
• Balancer, one of the leading DeFi automated market makers (AMM) for multiple tokens. Dive into their pools at https://balancer.finance/!
• Kraken, consistently rated the best and most secure cryptocurrency exchange, which can get you from fiat to DeFi
• Aave, an open source and non-custodial liquidity protocol where users can earn interest on deposits and borrow assets.

Do Kwon: When I first graduated from computer science, I'm a software engineer by training, I joined Microsoft as an engineer, and a lot of the work that I was doing was an NLP. So like when I first trained, maybe it was just like the department that I got placed in, but our team had 34 engineers. And there were, I think, at any given time, maybe three or four people didn't work, it was just not a very ambitious environment for a young aspiring engineer to be able to do lots of work. So I started to tinker with side projects, until I made a prototype that got me excited enough to quit and to make a startup.

“...at any given time, maybe three or four people didn't work, it was just not a very ambitious environment for a young aspiring engineer to be able to do lots of work.So I started to tinker with side projects, until I made a prototype that got me excited enough to quit and to make a startup.”

So what that technology did was that it was in Wi-Fi mesh, so it allowed user devices like cell phones and laptops to connect to each other via Wi-Fi direct or Bluetooth. And the idea was, even if you don't have direct internet access, you don't have a connection to a cell tower, or a Wi-Fi router, you could connect to the Internet through a network of other people around you. So it was sort of like a distributed network of human beings in densely populated or poor internet connection areas.

So we turned that into a sort of a B2B business, targeting large enterprises like airports and amusement parks and things like that. And then you can imagine that the standard Google search query for “distributed network” turns out a lot of things like Bitcoin and Ethereum. And then that's how I got started to get acclimated to crypto.

And a turning moment for me was when one of my friends invited me to a Facebook chat group with eight people, and it was mainly friends from college that were talking about things like Bitcoin, Ethereum, and different things like that. And then that conversation actually turned out to be a lot more intelligent than the stuff you find on Reddit. So that's when I got hooked and then started spending more time looking at whitepapers and going meetups and participating in community chats, until in 2017, I decided to quit my company, and then to do something in crypto full time.

Camila Russo: And so when you decided to quit your company, this is the distributed network, Wi-Fi solution that you started, right?

DK: Right.

Volatility as the Roadblock

CR: Okay. And then what was the first thing that you did in crypto full time?

DK: So I quit my job on Halloween of 2017, and then that day I went out for drinks with one of my friends who was running a small fund at the time. And then he invited another friend, and then that friend turned out to be the cofounder of Terra, Daniel Shin. And at the bar that day, he was looking to buy QTUM, which, I don't know if you remember QTUM, but the basic premise was that you combine the best of Bitcoin and Ethereum. And the webpage didn't really scale correctly and it was worth $6 billion.

And then I asked him, do you really want to invest in something that can't put together a website without typos or scaling correctly? And that led to an interesting conversation about what it would take to get a cryptocurrency that has lots and lots of real users. So that adoption conversation led to stablecoins and led to what is Terra today.

“And then I asked him, do you really want to invest in something that can't put together a website without typos or scaling correctly? And that led to an interesting conversation about what it would take to get a cryptocurrency that has lots and lots of real users.”

CR: That's so interesting. Because back in 2017, there weren't many stablecoins. Obviously, right now stablecoins are super popular, everyone uses them for trading. But in 2017, it was Tether, right? And I don't know if there were many others. DAI wasn't around, USDC wasn't around. What made you think of stablecoins as the mainstream use case?

DK: There were a couple of different reasons. But when most people were doing something in say digital commerce looked at Bitcoin for the first time, like the main roadblock to getting Bitcoin adopted wasn't so much speed or scalability, it was volatility. Right? So Bitcoin has upside, yes, but in a marginal business, like let's say, an ecommerce company, or any company that sells goods and services on the internet, it's actually cost prohibitive to have that volatility built into Bitcoin.

And Tesla can afford it because it's Tesla. But for most companies, it's really hard to run the risk of your proceeds going and dropping significantly in price, because that volatility can make it difficult to meet, let's say, payroll, or to pay distributors in a timely fashion. So from a business perspective, volatility was one of the greatest roadblocks to adopting cryptocurrencies. There's obviously custody, wallet management and different things like that. But fundamentally, if you couldn't get to a state where you could sort of guarantee crypto holding value stable to some mainstream unit of account, then we thought it'd be really hard to get it adopted.

“...from a business perspective, volatility was one of the greatest roadblocks to adopting cryptocurrencies.”

So there was this paper called Seigniorage Shares by this British gentleman called Robert Sams that I read through when I was doing research into crypto. And that actually later inspired a lot of algorithmic stablecoin projects, like Basecoin, Terra, and I think to some extent, MakerDAO and lots of other ones that came after it. And the idea was that, essentially, you would sort of have a second coin from the stablecoin absorb all the volatility from the first. And the idea was that there are a set of people that enjoyed that volatility for speculative purposes. And by having the second group absorb all the volatility from the first, you can allow the stablecoin to be used as a medium of exchange, store of value, and unit of account in day to day business.

How Terra Achieves Stability

CR: That’s interesting. So obviously that's the model that DAI has with well, at first, just ETH backing the stablecoin DAI and absorbing that volatility as people also take the other side. Right? With Terra, does it work in the same way? Like, how does Terra achieve its stability?

DK: Yeah, so in a weird way, MakerDAO is quite similar to Seigniorage Shares as well. Instead of creating their second currency, they simply use Ether. Right? So when, shall we say, the circulation of DAI is decreasing, it's as if Ether that's been locked up is being minted out of CDPs and entering the circulating market through liquidations and margin calls. And then when the markets are good, and there's more DAI in demand, it's as if more Ether is being burned and locked up into CDP contracts. So it's a very similar model where ETH absorbs the volatility of the DAI stablecoin in some sense.

So the primary difference between MakerDAO and Terra is that with Maker, it's an over-collateralization model. So given that they use a second larger currency, they don't have the luxury of, let's say, minting Ether whenever they choose to or burning Ether arbitrarily. So what that means is that any time that you want to create one unit of DAI stablecoin, you need to lock up, let's say $150 worth of ETH, because you want to leave some room for arbitrageurs to be able to make the system whole whenever you need to liquidate collateral to be able to buy back DAI stablecoins in circulation.

So for Terra, it's a different model where the second currency itself is native, and something that we've created. So the entire model is that any time that you want to mint a dollar’s worth of Terra stablecoin, you burn a dollar’s worth of LUNA. And vice versa, if you're looking to redeem a dollar’s worth of TerraUSD, the system mints and gives back a dollar’s worth of LUNA. So in that sense, given that it's a one-to-one peg between Terra stablecoins, and LUNA, it's more capital efficient. But of course, given that it's an entirely new currency, with black swan situations, you can make the argument that it is more fragile than what you would find in larger crypto backed counterparts like MakerDAO.

“So in that sense, given that it's a one-to-one peg between Terra stablecoins, and LUNA, it's more capital efficient. But of course, given that it's an entirely new currency, with black swan situations, you can make the argument that it is more fragile than what you would find in larger crypto backed counterparts like MakerDAO.”

Potential Stability Pitfalls

CR: So, without studying this very closely, or having a degree in economics, I'm thinking what the risk could be. Since it's easier to mint a lot of the stablecoin, can somebody buy a bunch of LUNA, and in that way mint a bunch of stablecoins? It seems like because there's no external currency to it, it seems like it's easier to game. How did you go about reaching some sort of equilibrium where it's expensive to come up with a bunch of LUNA and try to game the system?

DK: Yeah. So for all these stablecoin systems, the biggest point of risk is when cascading liquidations happen. So if you remember what happened on Black Friday, March of 2020, you had a bunch of DeFi protocols fail at the same time. So you had cascading liquidations where Maker protocol, for instance, tried to sell all the Ether that it had to be able to buy back DAI.

But there were too many transactions that were happening, transactions were clogged on the mempool. So you had a momentary situation where the Maker protocol was insolvent to $13 million or something. And it's exactly the same mechanism, right? So in a time where you're trying to sell the collateral in order to make the principle whole, similar to how if you're trying to sell a lot of ETH quickly, or let's say, a bunch of margin longs on Binance gets liquidated for BNB, and then it leads to like a 20-30% price shot really quickly, or any similar situation like that.

If a lot of people are trying to redeem TerraUSD quickly for LUNA, then in that case, LUNA gets sold off in the market very quickly and then that leads to sort of a cascading failure where the cost of redemption gets higher and higher. So I think the weakness there is that LUNA is going to be less liquid, and it's less valuable right now than Ethereum. So that cost of cascading failure is going to be larger. But the upside, though, is that given that the system doesn't keep any explicit collateral, if arbitrageurs were willing to wait until the cascading liquidations have stopped, then in that case you can still redeem TerraUSD for whatever market value of LUNA happens to be at that time. So cascading failures could happen if a lot of stablecoins are being redeemed at the same time. But the design of the protocol is that it should eventually be able to recover.

Adoption Leads to Stability

CR: Have you ever had an event like that happen?

DK: No, so we haven't had anything like that. So that's actually a really interesting point. So one of the things that I thought a lot of stablecoin designs are missing is that, at the end of the day, the core stability comes from utility and adoption. And what I mean by this is that the US dollar isn't the most stable because they have the most intricate monetary policy. In fact, the larger the economy, the less intricate the monetary policy. Right?

“...the US dollar isn't the most stable because they have the most intricate monetary policy. In fact, the larger the economy, the less intricate the monetary policy.”

The US dollar is the strongest because it is the most widely used fiat currency in the entire world. So taking inspiration from that, I thought, if you're going to be designing a stablecoin, and then present it to the wild, you're sort of hoping that some people pick it up for this use or another use. If it's poorly adopted, then it's going to be very hard to hold the peg. So if you saw an explosion of algorithmic stablecoins during summer last year, every single one of them fell through the peg, especially because they weren't able to foster any non-speculative use cases around them.

So basically, we thought, like a good way to make sure that Terra’s economy and its peg becomes robust over time is to create lots of use cases around them that are sticky, so for example, payments. So today, Terra’s stablecoins are being used by 2.6 million people in Korea for payment transactions. And it turns out that with payments, user adoption is fairly sticky. So once you start to get into the habit of, let's say, buying milk or your diapers with a payment medium, you generally don't switch for long periods of time. So if you're able to raise the demand over time by building and diversifying the use cases that are built around the stablecoin then the stability profile of the currency gets much, much stronger.

“So today, Terra’s stablecoins are being used by 2.6 million people in Korea for payment transactions. And it turns out that with payments, user adoption is fairly sticky. So once you start to get into the habit of, let's say, buying milk or your diapers with a payment medium, you generally don't switch for long periods of time.”

CR: And who's taking the other side? Who's using the LUNA side of it? Who's actually minting and redeeming these stablecoins?

DK: Well, it's profitable in lots of different cases for market making firms or trading firms to be doing the swaps. So for example, at any given time, if TerraUSD is trading above $1, then there's an arbitrage opportunity for people to buy up a dollar’s worth of LUNA, swap it to TerraUSD, and then sell it in the market for premium, and then vice versa for redemptions as well. So we have a number of trading entities and individuals that engage in these types of swaps, and probably many that we don't know about.

The Business Case For TerraUSD

CR: Alright, cool. Okay. So you mentioned that there's millions of people right now using the Terra stablecoins, and that's something that many cryptocurrency projects aspire to, obviously. What do you think you've done to get this right? How did you get this wide user base?

DK: Yeah. So to be completely honest with you, I think in the beginning, when we tried to sell the idea of crypto payments to ecommerce companies, I think most of them didn't get it. So I think at the beginning, Daniel’s sales network really helped because he had built the first and one of the largest ecommerce companies in the country, as well as the largest shared office, a WeWork copycat, that's about to IPO pretty soon. It is now bigger than WeWork in the country.

So he was able to get a lot of the ecommerce founders to agree to adopt the currency and then when we tried to sell it to their working level, there was very little understanding. But I think now there's a lot more clarity around the business proposition that's getting a lot of merchants to adopt the payment service, and then to push it very strongly to their consumers.

And the business case is pretty simple. Using stablecoin networks to settle has two major benefits. So first, you are able to settle very quickly. So, using traditional payment services, the average settlement time is about seven days. Which means that once a user pays on the ecommerce website for something, the payment gateway would enter a seven day delay for the money to be transferred to you, which is prohibitively expensive for a lot of different use cases.

So for example, if the platform works with lots of freelancers or smaller businesses, then in lots of cases, these smaller businesses need the money right away because they have working capital issues. So a cab driver, for instance, might need the money right away to be able to put food on their family's dinner table, or to be able to pay for gas for the next day’s rides. And the same for local small restaurants and things like that.

So by not settling using traditional settlement networks but to settle to merchants directly in stablecoin, we're able to cut down settlement times from seven days to six seconds, which is the average block time of the entire blockchain. And what we've also done is that we've added a lot of admin tools such that merchants can get analytics and understanding of how the revenues are coming from, swap out easily out to fiat if they choose to. So I think bringing that value proposition to merchants has been very, very valuable.

“So by not settling using traditional settlement networks but to settle to merchants directly in stablecoin, we're able to cut down settlement times from seven days to six seconds, which is the average block time of the entire blockchain.”

Second thing is, given that we don't use any of the existing settlement networks, we're able to cut down the fees from, I think the market rate is anywhere between 2.7-3.3%, we cut that down to 1.5%. So in a very tight marginal business, like ecommerce where, if you're not marketing, sometimes you're losing money or sometimes you're running on 2-3% margins. If you're able to save a percentage point, that's a big deal. So faster settlement times and cheaper fees combined, when we integrate with the merchant, they have a vested interest to make sure consumers use us instead of credit cards. So they do promotions and marketing events to make sure that people switch over.

“So in a very tight marginal business, like ecommerce where, if you're not marketing, sometimes you're losing money or sometimes you're running on 2-3% margins. If you're able to save a percentage point, that's a big deal.”

CR: That’s so interesting. And then do they swap the Terra stablecoins for local currency, for Korean won, or do they keep that?

DK: At some points, everyone swaps out to fiat because they needed to pay distributors. The speed at which people do conversion depends on industry to industry. So for example, for industries like fashion, they have a higher tolerance, because it's a high margin business, and they don’t need to swap very frequently. For very high velocity businesses, they swap very frequently. And that's because one of the main value propositions of this is that you're able to get your money quicker.

Why TerraUSD Has Worked

CR: Right. So why do you think that same go-to-market strategy hasn't really picked up that much in the US? To backtrack a little bit, in Korea, what percentage of merchants are using CHI, would you say?

DK: So, it's a very top heavy strategy. So we have about 45-50 very large merchants working with us. So for example, the largest convenience store chain, the largest bookstore. So it's kind of a premium brand name acquisition strategy. And then in terms of the long tail, it's not very heavy. But at 2.5 million, it's about 5% of the population here. So it's growing fast and fairly decent sized where we are.

CR: So what do you think the difference is?

DK: That's a good question. So I think actually, in crypto, the incentives are such that you can build something that's catered towards the crypto industry, and that's just going to be a lot more worthwhile than going out into the real world and trying to acquire users.

So for an average developer, I mean, I'm guessing that for most people, the goal is profit maximization. So if you wanted to get rich quickly, it's much easier to build a small team of people that are trying to, let's say, issue an NFT protocol or DeFi protocol, and then that's going to have a decent number of users, maybe a few thousand. And that's going to be enough to make the developers very wealthy, right, because of the market. So it doesn't make sense to go through the hurdle of educating consumers and then building relationships and then doing sales when you could probably 10X the results with 100 of the effort in a short time.

“...if you wanted to get rich quickly, it's much easier to build a small team of people that are trying to, let's say, issue an NFT protocol or DeFi protocol, and then that's going to have a decent number of users, maybe a few thousand. And that's going to be enough to make the developers very wealthy, right, because of the market.”

CR: I think that's a good hypothesis. So you started this in 2017, so in the middle of a bull market. But then all of the adoption and building continued through the bear market, do you think maybe that kind of played a role into why you moved really strongly into outside of crypto? Since it wasn't as easy as it is now to make money inside of crypto?

DK: So ultimately, I'm one of those people that believe that Wall Street only exists because of Main Street. At the end of the day, if you think about what are the key levers that move the economy, it's the people that are actually producing things with the capital that's been allocated. But if capital is allocated in a vacuum, then it's kind of meaningless, right, that's essentially what a bubble is.

“So ultimately, I'm one of those people that believe that Wall Street only exists because of Main Street. At the end of the day, if you think about what are the key levers that move the economy, it's the people that are actually producing things with the capital that's been allocated.”

It's not so much that I don't think that the things that are being done inwardly in crypto aren’t valuable, those are extremely valuable. But I felt like there was nobody that was trying to do things looking outwards into growing the pie, bringing more users in and making crypto more approachable for the everyday user. So somebody had to do the job and I felt like that was as good a place to start as any. The jury's still out as to whether that was a good idea.

But it's extremely fulfilling. Because you see, for example, even outside of payments, if you look at Mirror protocol, the thesis is that you can invest into any asset class at once without restrictions or hampering by existing systems. And our top markets are Thailand, China, and Indonesia, where the access to US equities is really, really difficult to do. And we're seeing very small ticket sizes come in, like maybe there’s a $50 stock purchase. I've seen even a $9 purchase.

So obviously there's something that we're doing to lift barriers to allow smaller users to come in that are outside of crypto today and then the value proposition is making their lives better. It feels good when you see stuff like that.

“So obviously there's something that we're doing to lift barriers to allow smaller users to come in that are outside of crypto today and then the value proposition is making their lives better. It feels good when you see stuff like that.”

Shipping Too Quickly

CR: 100% I agree. So do you think right now, is DeFi in a bubble in that sense? Are DeFi protocols building inwardly and just making money as a cycling money system of yield farming and pump and dumping new tokens and farms? It does feel a little bit like a bubble to be honest, this, in a word, “cycling” of capital, what do you think?

DK: But Camila, if you think about it, in 2017, the problem was that teams were shipping too slowly. Right? So you've seen companies raise $100 million, and then ship nothing two years later. But with DeFi, it was sort of the opposite problem where people weren't raising any money, and then shipping too quickly and then things breaking.

“...in 2017, the problem was that teams were shipping too slowly. Right? So you've seen companies raise $100 million, and then ship nothing two years later. But with DeFi, it was sort of the opposite problem where people weren't raising any money, and then shipping too quickly and then things breaking.”

So obviously things are moving for the better, in the sense that, even if you look at private markets, valuations for new DeFi protocols that are coming out are much, much slower. There's no more companies that are raising $50 million anymore. So in some sense, I would say that there is a lot of hype, but I think the fundamentals in DeFi are a lot stronger than what ICOs looked like in 2017.

And I think it's fine if there's lots of different projects like food coins, like Pickle and FrenchFry, or Soufflé, these are probably real coins, right? It's fine if these assets are yield farming in a vacuum, because if there's projects like Mirror and Anchor and different types of things bringing in users to fill in the base layer, then in that case, the more speculative stuff that gets built out on top just leads to better allocation of capital. So I'm trying to do the former such that the water doesn't turn into a bubble.

CR: This is another world compared to 2017, there are actual useful projects being built, actual users, even though it feels like much of it is speculation. I agree that the outcome of all this speculation will be better products and bringing in other types of users who are, hopefully, leveraging open finance for their real lives and not just for flipping some food coin. Okay, so with this very large user base and people actually using the Terra stablecoin for payments, how is the Terra blockchain able to support this level of transaction and throughput, how is it built?

Terra’s Architecture

DK: So Terra is a deep PoS consensus mechanism with LUNA as the staking token. So we use Tendermint and the Cosmos SDK. So right now, there's a lot of low hanging fruit in terms of performance optimization that needs to be done in Tendermint core or the SDK itself. So right now, I would say, the practical throughput that we're seeing is maybe 700 TPS. But I think with a few optimizations, you can get to a couple thousand.

And this leads to an interesting dialogue about Cosmos and how I look at Tendermint in general. But I think actually, the branding of Cosmos as an interoperability solution is not a great one. I think it should actually be branded as a scaling solution.And if you think about it, if you look at all the different Cosmos zones that are built in the system, they're sort of exactly the same as shards in a sharded blockchain except that they're branded independently. They run according to a very similar set of rules. They just run local computation that are relevant to that side of the world. Right? And then you sort of have IBC as the consensus layer, where things need to be sorted to global state, then you sync up using relars.

“I think actually, the branding of Cosmos as an interoperability solution is not a great one. I think it should actually be branded as a scaling solution. And if you think about it, if you look at all the different Cosmos zones that are built in the system, they're sort of exactly the same as shards in a sharded blockchain except that they're branded independently.”

So, I think the future of Terra, using this technology is pretty simple. It's essentially going to get to a branded, sharded blockchain, where each of the applications that are being built out, for example, Chai could be its own independent state machine, and then maybe even within Chai, depending on geographic regions, things like that could have its own state machines. And then Mirror protocol can have its own state. Anchor can have its own state. And wherever it becomes important to settle to the monetary layer, which is to use Terra stablecoins, then in that case you sync global state to a global Terra hub using IBC.

CR: Okay. And so DPoS is delegated proof of stake, right?

DK: Yes.

CR: And that's the consensus mechanism of the Cosmos ecosystem. Is that right?

DK: Yeah. So all Cosmos related projects use a proof of stake consensus mechanism called Tendermint, which is sort of like the core innovation of the Cosmos ecosystem. And I would say that for most intents and purposes, it’s the first proof of stake mechanism used widely in blockchains.

CR: Right. And then when you achieve, finality, like when all the transactions are confirmed, and they are downloaded to the Layer 1, does that happen on the same blockchain that all of the Tendermint blockchains use or is it like Terra specific?

DK: No, they’re all separate. They're all separate launches.

CR: Oh, they're all separate shards?

DK: Well, right now, they're all separate blockchains. And then the Cosmos vision is that they will tie together all these separate blockchains, and then have some sort of global hub. And the technology that enables this is called IBC. But I don't think any of the Cosmos blockchains have connected using IBC yet, so they're all independent, running their own thing.

CR: Okay. And the reason why you would want to connect using IBC is because then you can easily transfer assets from one blockchain to another in this Cosmos ecosystem?

DK: No. So that's what I mean when I say that is the wall narrative. So Cosmos as an Interoperability solution, I think, is valuable, but I think it's far more valuable to think of it as a scaling solution. Because if you have the branded. logically independent blockchains or independent shards, then in that case, you can separate logic into all these different shards and then just only keep global state for the really important stuff. Yeah.

Blockchain Trilemma

CR: Okay. So in this blockchain trilemma that's famously decentralization, security, throughput / performance, where do you think that Terra is compromising? Seems like it might be decentralization, like how many nodes are there in this delegated proof of stake blockchain?

DK: Right now there's 100.

CR: Okay. Yeah. Do you think that's where you're compromising to achieve greater throughput?

DK: Yeah. So I would say there's definitely sacrifice in decentralization. But every time where state becomes too much to handle in one set of validators, you need to split off into different security consensus. So in some sense, in order for the blockchain to scale, we've chosen a design whereby we split off security into local zones or local hubs of security. So in some sense, as the blockchain is scaling, there's a smaller set of data that the global consensus reaches agreement on.

“I would say there's definitely sacrifice in decentralization. But every time where state becomes too much to handle in one set of validators, you need to split off into different security consensus. So in some sense, in order for the blockchain to scale, we've chosen a design whereby we split off security into local zones or local hubs of security.”

CR: Okay. Let me see if I understood this. So when the state of the blockchain becomes too big because there are too many transactions, you split up that state, like that information, into separate blockchains or into other notes?

DK: So we haven't done this yet. And right now, everything is running on the Terra blockchain. But the eventual goal is that if it gets too crowded, we'll just spin up another blockchain using Tendermint and the Cosmos SDK, and then we'll connect that to the Terra hub using IBC. And in some sense, if you have too many of these smaller local blockchains, and they have their own validators, you could lead to a situation where it's less secure than the main blockchain, if that makes sense.

CR: Right. Okay. So that's the risk of doing that. But you haven't had the need to actually do that yet?

DK: Right.

Anchor Protocol

CR: Cool. Okay. And then changing gears a little bit. I wanted to talk about Anchor, this new lending protocol that you launched recently, last month, right?

DK: Yes.

CR: And everyone should head to our YouTube channel to watch the tutorial we did on Anchor. So we'd love to hear more about this. Like, what's the idea behind Anchor? How is it different to other DeFi lending protocols? And importantly, value prop here was stable yields, so how are you able to offer that?

DK: Yeah. So to talk about the motivation for the project a little bit, I think we're living in an interesting time, where interest rates are kept artificially low for political purposes. So if you look at how inflation is changing against most CPI, the dollar is losing value to the tune of 2-3%. But if you look at more competitive indexes, like for example, the S&P 500, or most real estate indexes, it's losing value to the tune of almost double digits.

So we're at a point in time where the average household is not making enough money from savings to justify cost of capital. Obviously, that's not good. So where Anchor wants to get to is the ability to offer an interest rate that is apolitical and unbiased, and really justifies the cost of holding money in a savings account for the average household. How it differs from sort of day to day hustle DeFi protocols like Aave or Compound is that if you look at, let's say, Aave or Compound, for instance, the interest rate that you can get a stablecoins fluctuates pretty widely anywhere between 0-12%, in a matter of hours. And the reason why it fluctuates is because the only reason to borrow USDC or stablecoins from these protocols is demand from leverage on speculation.

So when demand for leverage is high, then yield is going to be high, and then when that opportunity diminishes, then yields are going to collapse to zero. So obviously, that's not a great savings experience for the user. Because when I'm putting money in a savings account, I kind of want to put it there and forget about it. But if I'm not sure if I'm going to get 0% or 3% or 7% while running technical risk of putting money in a smart contract, then it just doesn't make sense for me to do.

“What Anchor wants to get to is the ability to offer an interest rate that is apolitical and unbiased, and really justifies the cost of holding money in a savings account for the average household.”

Anchor is able to offer a 20% fixed income product for the user. So the idea is that if you deposit TerraUSD into Anchor, you're able to earn 20% per annum and then that rate is guaranteed, unless the rate is changed by governance. So how this works is, whenever you deposit stablecoins into Anchor, a portion of the deposits are facilitated to loan out to users that borrow stablecoin by locking up their PoS collateral in a smart contract. And then the protocol puts that PoS collateral to work to earn staking yield across multiple different PoS chains, and then confers that PoS yield to the depositor in the form of yield.

So in some sense, like by using cash flow based assets as collateral in Anchor’s money market, you're able to get to a state where you have a diversified stream of proof of stake yields coming in from multiple different blockchains powering a stable yield for the user. So in some sense, it's a lot more robust than what you would see from DeFi yield farming schemes or any other source of yield in blockchains today.

Maintaining Yields

CR: Super interesting. But that depends on proof of stake yields remaining high. Right? How can you guarantee that proof of stake yields will remain at 20%?

DK: Well, to do a little bit of the back of the envelope, it's whatever the average proof of stake yields are times the minimum collateralization ratio, because the loans are going to be, let's say half of whatever outstanding collateral is out there, because the system is over-collateralized.

So, for example, if the average staking yield is, let's say 12%, and the minimal collateralization ratio is 200%, then the system can technically offer 24% yields without the system going bankrupt. So, as to whether PoS staking yields are going to stay at 10-12% range for the foreseeable future, most blockchains have not fixed monetary policy or fixed token emission: which means that denominated in that PoS coin, it's going to yield that fixed number of points for a really long period of time. So some of them have an expiration date on the emission, some of them keep going in perpetuity. But, yeah, given that the monetary policy is fixed, it's going to keep going for some time.

I think, though, that as blockchains mature, PoS blockchains are going to try to curb down emissions to a level that is more manageable, because otherwise, it's not great and it’s too dilutive. But I think that's going to take many, many years for that to happen. And then when it does, I suspect that Anchor rates will have to keep coming down as well. But the point is that it's never going to get to a state where it's lower than inflation, if that makes sense.

Multi-chain Integration

CR: Yeah, super interesting. And so, which proof of stakes chains are you using for this?

DK: Yeah, so right now, when we first launched last month, we only integrated with LUNA. Upcoming chains are the staking derivative for Ether that we're going to integrate with, and then Solana, which is also going to be interesting. And then we're also going to be working on Polkadot and then Cosmos, and yeah, a few other ones for which we don't have solid development timelines for yet. But the goal is to basically encompass most large-cap PoS coins.

“We actually are setting up Anchor contracts in each of the chains. So there's going to be a version of Anchor on Ethereum. There's going to be a version of Binance Smart Chain. There's going to be a version of Solana and that sort of support. And then you would even be able to composite in other stablecoins.”

CR: So right now, it's all kind of staying within the same Terra ecosystem. Like you're using LUNA, the Terra blockchain as the back end for Anchor to get these proof of stake yields. But when you branch out to other proof of stake blockchains, you'll have to convert those. I mean, it seems like there will be more steps involved in Anchor, you'll have to convert those other coins, outside coins to Terra, and then bring that yield to Anchor users, right?

DK: So that's really interesting that you asked that. We actually are setting up Anchor contracts in each of the chains. So there's going to be a version of Anchor on Ethereum. There's going to be a version of Binance Smart Chain. There's going to be a version of Solana and that sort of support. And then you would even be able to composite in other stablecoins. So for example, you'd be able to deposit in Tether Circle, even DAI. And then the idea is that once a user deposits Tether, we're going to convert that into USD using Curv, and then that USD is going to be carried over to the Terra blockchain such that it can be deposited in Anchor and then saved for collateral as well. So it's going to be a blockchain native experience where you might be playing around with this.

CR: Oh, that's interesting. Okay. So your idea is to build these Terra protocols across chains?

DK: Yeah.

CR: Okay. And so in that way, whenever you're using Terra, if you're on Ethereum, you don't have to go somewhere else to use Anchor, you can just use it on Ethereum?

DK: Got it. Yeah, yeah, that's the idea. So I mean, we sort of think about running Terra like running a very small micro-developing economy. So we sort of think about the activity that happens using Terra stablecoins outside of the blockchain as foreign exports. So, the more that our products get used, then integrated into places like Ethereum, and Binance Smart Chain, then it's good for the economy.

And it actually turns out that stablecoins, once they cross over into a different ecosystem, they don't often come back. So in some sense, in terms of guaranteeing stability of the overall ecosystem, it's even better than things that are happening in house on the Terra blockchain. So that's an interesting observation.

Mirror

CR: Will you do the same with Mirror, like build on different blockchains?

DK: Oh, yeah. So Mirror is actually already on Uniswap, on Ethereum, and PancakeSwap on Binance Smart Chain. And I don't know if this is still true now. But when, maybe just even about a month ago, Mirror pools with USD were some of the most liquid on PancakeSwap. It's probably changed at this point, but that was kind of interesting to see.

CR: Can you talk more about the motivation behind Mirror?

DK: Yeah. So when I first started thinking about Mirror, I think it was about September of last year. And then it was right after DeFi summer, and some of that was still going on. But I thought it was really funny how all the points were naming themselves after coins. And then I saw like rug pull after rug pull, and then it was entertaining, but I saw a real opportunity to be able to build DeFi primitives that have value tied to real things, and that's pretty much what Mirror tries to do.

So Mirror is basically a synthetics protocol, where you can issue tokens that check the price of anything. So it could be Tesla stock, Apple stock, even GameStop, or Coinbase is going to be listing on Mirror within a few days. But basically, the idea is that you're going to have coins that track the price of every asset in the real world.

And then I was thinking that for, let's say, like a derivatives platform, a DeFi platform that's freshly launching, it would make sense to get people that are holding near the assets of US equities, and then airdropping governance tokens to people that do this. So if you had assets to the real world value with much lower risk of impermanent loss, these would be very attractive tools that DeFi developers can build on to create things with more tangible value, if that makes sense.

“But the highest levels of pickup that we've seen is in the markets like Thailand and China, and Indonesia, where access to US equities is fragmented and difficult.”

And then it turns out that, from a very basic use case, there's lots of places in the world where they don't have quality access to US equities. So US companies lead the world in terms of innovation, but it's in a sense to reason that their assets are very highly desirable. But there are lots of places where you can't get good access to US equities. Or even in cases when you do, local adversarial regulations make it very difficult for you to be able to access it early.

So for example, in Korea and Japan, capital gains tax on single stock on US equities is 26%. Whereas on domestic equities it’s 0.3%. And every time that demand for US equities goes up, the financial regulator issued a statement saying, hey, look, we're considering hiking up taxes again. So it's fundamentally designed to prevent capital flight and to keep money in domestic equity exchanges as much as possible.

So as I alluded to previously, what's been really interesting is that we launched our Mirror protocol on Twitter and we've done no localization to other languages besides English and have done no marketing efforts whatsoever. But the highest levels of pickup that we've seen is in the markets like Thailand and China, and Indonesia, where access to US equities is fragmented and difficult. So I mean, we've sort of been very surprised ourselves at the level of organic pickup play out in markets where we thought it would have the strongest uses. That's been very interesting to see.

CR: Yeah. I love the idea behind Mirror, just enabling access for US stocks for everyone. I think people in the US, assume that everyone can buy Tesla stock, but that's obviously not the case. How is this peg actually achieved? Like, do you have to hold actual Tesla stock and issue synthetic assets against it, like, how does it technically work?

DK: Yeah, there's no centralized counterparty. This is kind of like an inverse version of MakerDAO. So if you look at MakerDAO, you have collateral that's posted in ETH and then a stable synthetic that's issued, so that's the DAI stable coin. In the Mirror system, it's kind of the opposite. So you have a stable collateral which is TerraUSD stablecoin and then you issue several different oracle synthetics. So basically, you have a pool of stablecoin collateral that's guaranteeing the principle for these oracle synthetics. And whenever the price of these stocks goes up too high, then in that case, the stablecoin, collateral is liquidated to be able to buy back these synthetic assets.

CR: And then their prices kept via an oracle, I guess that talks to the actual price of the stocks?

DK: Right.

Mirror vs Synthetix

CR: On Mirror, how has adoption been so far? And also, how is it different from something like Synthetix, which was kind of the original Synthetic protocol?

DK: Yeah. So to talk about the pickup first, we're seeing about 20,000-22,000 daily active users on the Mirror web app. On the Mirror mobile app, we don't run that service, but should have to check. But also, growth there has been pretty interesting. There's about $2 billion in TVL locked up across Uniswap, TerraSwap, and then PancakeSwap, so across all these three different exchanges.

We're seeing anywhere between $50-100 million in daily trading volume. And it's picking up a pretty fast pace over time. So, yeah, all of that has been pretty interesting.

In terms of the difference with Synthetics is, if you look at Mirror, it's a very simple value proposition. So the idea is that you buy Mirror, the assets, and you hold them, and then you can exit whenever the price of Mirror assets goes up. So in terms of buying and selling these assets, the mathematics are pretty simple. You hold equities, they go up in value, you exit, you've made that money.

“So that's a tradeoff that you make with Mirror, you have to have explicit liquidity. You need to put LP into Uniswap and Terraswap and PancakeSwap, and then that liquidity is bounding. So, if the trades are too large for the liquidity, costs are going to go up a lot.”

With Synthetics, there's a couple of points where things are made a little bit difficult. So the use of schema whereby users have to first mint SUSD and then use SUSD to be able to trade these various different Synthetics. And then they have this idea of the infinite liquidity, which means that you can swap SUSD to any Syn without paying a large spread, just maybe a fixed 0.1% or 0.25% in terms of fees.

But the problem there is whenever you're looking to exit, you're not actually trading against the stock market. So if Apple goes up 20%, you haven't made 20%, you're actually trading against other Synthetics traders in the ecosystem. So if all the stock prices start to rally, and then you haven't traded correctly, and then everybody else sort of trades well, and then they've sort of made a similar set of trades, you could end up in a situation where you actually don't make that much money, even if the stock market rallied significantly.

So that's a tradeoff that you make with Mirror, you have to have explicit liquidity. You need to put LP into Uniswap and Terraswap and PancakeSwap, and then that liquidity is bounding. So, if the trades are too large for the liquidity, costs are going to go up a lot. On the other hand, with Synthetics, they have infinite liquidity, so you'll be able to make 100 million trades without suffering significant slippage. But the cost in this is that you’re trading against other traders. So your return profile is going to be a little bit more complicated than what you would see in the more [inaudible 55:33], if that makes sense.

CR: Got it. And then, to centralize on liquidity, or liquidity providers backing these Synthetic assets on Mirror, you’re getting trading fees on people buying these Synthetic assets?

DK: Right. So, I think on PancakeSwap, these might be a little bit lower, but the liquidity providers make 0.3% from every trade, as well as the liquidity mining program going on with the MIR Governance token, which is also pretty [inaudible 56:15].

Terra Innovation

CR: Okay, cool. Okay. And then to wrap up, I know, we're running out of time. There's a lot going on in the ecosystem and so many different interesting protocols in the Terra blockchain. But what's your bigger vision, how do all these pieces fit together? And where does Terra fit into the broader DeFi?

DK: Yeah. So if we run over time, and you have to cut something, I’m probably not the person that you should cut. So the way that I think about it is that all the Layer 1s sort of play in the cloud or the computation category. And the idea is that if a lot of people use this distributed computer, then the underlying token is going to grow in value as a commodity to be able to pay for bandwidth on this computer.

So Terra is a little bit different. We do have an execution layer, we do have lots of different apps that are coming out. But we're not in the business, or we're not in the game of thinking of ourselves as the fastest execution engine or the best distributed computer. Our main products, our only product is the money that we've created, which is the Terra stablecoin. And all the different products that we've created on top of it, like CHI, like Mirror, like Anchor are simply features to make this money more useful.

“What's been really interesting is that the first generation of Terra protocols has been mainly built by me and Terraform Labs. But the second generation and I think a lot of these might be even better than the first, are being built by community members that just stumbled across the protocol in some way, and got really excited by it.”

So the idea is that as we have a savings protocol that makes Terra stablecoin more attractive to hold, we have Mirror that makes Terra stablecoin the primary gateway to be able to invest. We have CHI to make our stablecoins easier to pay and lots of different protocols that are happening. The Terra stablecoin is going to be fundamentally more useful to use than either the US dollar or DAI or any other stablecoin that might come after it.

And the great thing about this is, as Terra stablecoins get more widely used, there's a linear relationship in terms of how the LUNA token grows in value. So it's not so much that lots of Terra is minted so LUNA is going to moon or anything like that, but having an asset capture value in a very succinct way and the growth of a stablecoin, I think is one of the great ways to rally a community around. And these community members, once they're incentivized through this sort of value accrual mechanism, sort of roll up the sleeves and go out and build something.

So what's been really interesting is that the first generation of Terra protocols has been mainly built by me and Terraform Labs. But the second generation and I think a lot of these might be even better than the first, are being built by community members that just stumbled across the protocol in some way, and got really excited by it. So that's been humbling to see.

CR: Very cool. Okay. So I guess the bigger vision is to make the Terra stablecoin a more useful currency than even the US dollar.

DK: Right.

DeFi Meets TradFi

CR: Very cool. Awesome. And then, just finally, your vision for DeFi or Web 3.0, is it taking over everything, is it sharing the spotlight with TradFi? What do you think?

DK: Yeah, I think the distinction between TradFi and DeFi might be thinner than most people think. So for example, I think I sit in the intersection of both of these things a little bit, like, for example, my payment services are definitely TradFi, like a lot of the DeFi protocols that have been created are very, very DeFi. But I think the holy grail is interesting things that happen when these two things sort of start to interplay with each other. And at the intersection, that could be really interesting things.

“I think the holy grail is interesting things that happen when these two things sort of start to interplay with each other.

So for example, if we had Anchor yields enabled across FinTech apps, like for example, CHI, or let's say, Venmo, how amazing would that be? Like, all you need to do is in your favorite app that you use, to send money to your friends, your money while sitting idle is earning 20%. So that's amazing. And like on Robinhood, well, actually, Robinhood is a bad example. But imagine a world where like something that looks like Robinhood and doesn't have censorship possibilities is propagated to everyone in the world, even if you're not in the United States and anybody, be they in Malaysia, or let's say Vietnam, could get easy access to buy things like Tesla and Apple whenever they want it.

So I think the vision is you should have a very large acquisition funnel of TradFi powerhouses that are trying users into very basic use cases of DeFi like savings and synthetics, and different things like that. And there's always going to be a more sophisticated version of DeFi, this would be sort of our version of Wall Street, where people play around with perpetual swaps and normal financial instruments. And I think, in both of these different areas of DeFi, there's going to be tremendous growth. And I think it's going to be more symbiotic than what most people realize.

Share The Defiant

The Defiant is a daily newsletter focusing on decentralized finance, a new financial system that’s being built on top of open blockchains. The space is evolving at breakneck speed and revolutionizing tech and money. Spread the word and share!